A Challenge from Chemistry
While working on optimal control at the Zuse Institute Berlin, Michael Wulkow was introduced to a new idea by Prof. Peter Deuflhard: simulate polymer design on the computer. The challenge? Bridging the gap between mathematical methods and chemical intuition. Early meetings with Prof. Reichert’s group at TU Berlin made one thing clear — chemists and mathematicians spoke very different languages.
Computing in Technology since 1992
In 2022, CiT celebrated its 30th anniversary. The following timeline shows how the company has developed over the last three decades and beyond.
1988
Foundations in Math and Modeling
The collaboration led to two key publications in 1989, introducing the weighted discrete Galerkin method — an early numerical approach for simulating polymerization. While the use of weight functions would later be set aside, the mathematical framework built around orthogonal polynomials laid the foundation for what would become the h‑p-algorithm in PREDICI.1989
1990
A Theoretical Breakthrough
In autumn 1990, Michael Wulkow submitted his PhD thesis, Numerical Treatment of Countable Systems of Ordinary Differential Equations. Still based on weighted functions, it also employed a novel time-integration approach — a key step that would later make the h‑p-algorithm possible.1992 – The Birth of the h‑p‑Algorithm
The first version of the discrete Galerkin h‑p method took shape in 1992 — a more flexible, modular approach that would become the core of PREDICI. Developed on a Mac IIse, this early prototype already featured live graphics during simulation. Object-oriented programming, still rare at the time, was essential to the method — and the Mac made it possible.1992
1992
1992 – CiT Is Founded
In parallel with the early development of the h‑p method, Computing in Technology (CiT) was founded near Oldenburg, in Lower Saxony, Germany.The First PREDICI Is Delivered
In 1993, the first full version of the software — then called MEDICI — was delivered to TU Berlin and a major German chemical company. In response to user needs, development moved to Windows 3.11 using Borland C++, one of the first compilers with full template support. The core structure built then, based on modern C++ and modular design, still underlies today’s PREDICI — continually expanded, but never outdated.1993
1995
1995 – From Chains to Particles
As CiT’s focus expanded, so did the underlying mathematics. In 1995, the core ideas behind the h‑p algorithm were extended from discrete chain lengths to continuous particle size distributions. This led to the development of PARSIVAL — a solver for complex population balance equations in crystallization, precipitation, and particle systems, using integrals instead of sums but rooted in the same rigorous approach.1997 – Expansion to General Kinetics and Other Applications
In 1997, CiT introduced PRESTO-KINETICS, extending its modeling approach to general reaction systems, including catalysis and reactor design. Additional tools followed in areas such as pharmacokinetics, process observation, and environmental systems.1997
2002
2002 – Leadership and Expertise Expand
After several years at CiT, Regina Telgmann became co-manager in 2002. With a background in mathematics and software development, she would go on to complete her PhD in 2007 with a thesis on computer-aided modeling and model discrimination.2002–2006 – Expanding into New Applications
In the early 2000s, growing user demand led to a series of ambitious projects that pushed the boundaries of CiT’s tools. These included coupling CFD with population balances, building real-time simulation platforms, developing a polypropylene training simulator, and modeling bio-filters in 2D. During this period, publications based on PREDICI also increased significantly — reflecting its growing role in both industry and research.2002
2008
2008 – Modeling the Verbund
CiT contributed to the development of a site-wide simulation tool for BASF’s largest integrated production site, modeling material and energy flows across the Verbund network of interconnected plants. While fundamentally different from polymer or particle models, the project required innovative modeling strategies and highly efficient algorithms to solve large-scale, rule-based systems — often involving over 20,000 variables in near real time. The approach is described in a scientific publication and remains in active use today.2010s – A New Generation of Predici
Starting around 2010, CiT began developing a fully restructured version of PREDICI — introducing a new compiler platform, modern interface framework, and unified project environment. The goal was a completely new user experience: multiple models, flexible navigation, and parallel workflows. Despite the major redesign, the numerical core remained unchanged, ensuring identical results and full compatibility with older models. The hybrid Monte Carlo solver was added during this phase, alongside extensive testing to preserve trust and continuity. Today, most users work exclusively in this newer environment — often unaware of the legacy it replaced.since 2010
2015
2015 – First Release of the New Predici
By 2015, the restructured version of PREDICI was ready for delivery. With its modular architecture, unified project environment, and expanded simulation features, it marked a major leap forward in usability and flexibility — while preserving full numerical consistency with earlier versions. Long-time users found their models fully reusable, with results unchanged.2018 – Publication of the Modeling Textbook
Together with Prof. Dr. Klaus Hungenberg (formely BASF), Michael Wulkow published Modeling and Simulation in Polymer Reaction Engineering — a textbook that reflects decades of collaboration between industry and mathematical modeling. Rather than a software manual, the book presents the underlying structures, balance equations, and logic of modular modeling. It remains a reference for those building models beyond ready-made templates.2018
2021
2021 – Modernized Look and Feel
In 2021, CiT updated the PREDICI interface using the DevExpress platform. The result was a more intuitive, professional design with a layout and interaction style familiar to users of modern Office software.2022 – Integration with KBC’s Petro-SIM
In 2022, CiT began a collaboration with KBC to integrate PREDICI into their Petro-SIM flowsheet simulator. Using OLE-based communication, a dedicated interface was developed that allows PREDICI models to function as polymer unit operations within full process plant simulations. This brought detailed polymer reaction modeling into large-scale flowsheet contexts — with a special focus on thermodynamic coupling and process integration.2022
2023
2023 – 30 Years of PREDICI
Three decades after its first release, PREDICI remains an actively developed and widely used tool for polymer and kinetic modeling. The 30-year mark reflects not just longevity, but continuity — consistent numerical foundations, ongoing improvements, and a user base that continues to grow across generations of researchers and engineers.Today – Still Evolving
CiT continues to develop and refine its software in close connection with both academic research and industrial needs. Current work includes advanced reactor cascade and tubular reactor modeling, new levels of scripting flexibility, structured support for vectors and matrices, and a powerful placeholder system for automated copolymerization setup. The goal remains the same: to make sophisticated modeling accessible, adaptable, and reliable — across fields, processes, and generations.2025
Where to find us:
Harry-Wilters-Ring 27
26180 Rastede | Germany
Email: m.wulkow [at] cit-wulkow [dot] de
+49 (0) 44 02 / 84 2 48
LATEST NEWS:
- Upcoming Keynote and Workshop at Polymer Reaction Engineering XII7. April 2025 - 10:52
- CiT continues partnership with KBC7. April 2025 - 10:33
- New Predici11 version 11. Maxwell available7. April 2025 - 10:28
- Free-Radical Homopolymerization Kinetic7. April 2025 - 10:26
- Multiscale modeling of the microbial production of polyhydroxyalkanoates7. April 2025 - 10:25